The Journal of Infectious Diseases · 2011 · 19 citations · 33 references
Veterinary MedicineInfectious Disease ControlInfectious DiseasesClinical EpidemiologyEpidemiologyInfectious Disease EpidemiologyDisease EpidemiologyInfectious Disease Risk FactorsInfectious Disease SurveillanceMosquito-borne DiseasesRisk Factor EpidemiologyViral EpidemiologyBayesian Likelihood EstimationCase-control Study DesignFull LikelihoodMosquito-borne Disease EpidemiologyMosquito-borne Disease RiskMultivariable Regression AnalysisPopulation-based Case AscertainmentSindbis Virus Infection
BACKGROUND: Sindbis virus (SINV) is an arthropod-borne alphavirus that causes rash and arthritis. In Finland, epidemics occur cyclically, but factors associated with clinical SINV infection are largely unknown. We conducted a population-based case-control study during the epidemic year 2002. METHODS: SINV cases were serologically confirmed and reported to the National Infectious Disease Registry. Five control subjects, matched for age, sex, and residence, were selected from the National Population Information System. Data were collected using a self-administered mail survey. Conditional logistic regression models were used to identify independent risk factors; missing data were addressed using Bayesian full-likelihood modeling. RESULTS: A total of 337 case patients (58% female; age range, 1-94 y) and 934 control subjects were enrolled. Reported exposure to mosquito bites (matched odds ratio [mOR], 16.7; 95% confidence interval [CI], 9.1-33.4) and spending time in woods or marshland (mOR, 1.8; 95% CI, 1.3-2.5) were independently associated with SINV infection in the multivariable model. The population-attributable risk for mosquito bites was 87.2%. There were dose-response relations for increased number of insect bites (mOR, 23.8-72.5) and increased time spent in woods or marshland (mOR, 1.3-2.2). CONCLUSIONS: Educating the public in endemic areas to avoid mosquito exposure and use protective measures remain important prevention measures for SINV infection.
33
Missing data: Our view of the state of the art.
Joseph L. Schafer, J. A. Graham · Psychological Methods · 2002 · 10.9K citations
Bayesian Statistical Modelling
Peter Congdon · Measurement Science and Technology · 2002 · 683 citations